Universidad Diego Portales

Escuela de Informática y Telecomunicaciones Av. Ejército 441 Santiago Región Metropolitana Chile +56 (2) 676-8121 diego.dujovne@mail.udp.cl

Sandelman Software Works

mcr+ietf@sandelman.ca

Internet 6TiSCH BRSKI enroll zero-touch DODAG balancing LLN balancing In the Time-Slotted Channel Hopping (TSCH) mode of IEEE Std 802.15.4, opportunities for broadcasts are limited to specific times and specific channels. Routers in a TSCH network transmit Enhanced Beacon (EB) frames to announce the presence of the network. This document provides a mechanism by which additional information critical for new nodes (pledges) and long-sleeping nodes may be carried within the EB in order to conserve use of broadcast opportunities.

Introduction describes the use of the Time-Slotted Channel Hopping (TSCH) mode of . In TSCH mode of IEEE Std 802.15.4, opportunities for broadcasts are limited to specific times and specific channels. Routers in a TSCH network transmit Enhanced Beacon (EB) frames during broadcast slots in order to announce the time and channel schedule. This document defines a new IETF Information Element (IE) subtype to place into the EB to provide join and enrollment information to prospective pledges in a more efficient way. The following subsections explain the problem being solved, which justifies carrying the join and enrollment information in the EB.

Terminology The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 when, and only when, they appear in all capitals, as shown here. Other terminology can be found in .

Layer 2 Synchronization As explained in , the EB has a number of purposes: it carries synchronization information such as the Absolute Slot Number (ASN) and Join Metric and identifiers for the timeslot template and the channel hopping sequence, and it indicates the TSCH slotframe. An EB announces the existence of a TSCH network and the nodes already joined to that network. Receiving an EB allows a Joining Node (pledge) to learn about the network and to synchronize with it. The EB may also be used as a means for a node already part of the network to resynchronize . There are a limited number of timeslots designated as broadcast slots by each router in the network. Considering 10 ms slots and a slotframe length of 100, these slots are rare and could result in only 1 slot per second for broadcasts, which needs to be used for the beacon. Additional broadcasts for Router Advertisements (RA) or Neighbor Discovery (ND) could be even more scarce.

Layer 3 Synchronization: IPv6 Router Solicitations and Advertisements At Layer 3, defines a mechanism by which nodes learn about routers by receiving multicast RAs. If no RA is received within a set time, then a Router Solicitation (RS) may be transmitted as a multicast, to which an RA will be received, usually unicast. Although reduces the amount of multicast necessary for address resolution via Neighbor Solicitation (NS) messages, it still requires multicast of either RAs or RSes. This is an expensive operation for two reasons: there are few multicast timeslots for unsolicited RAs; and if a pledge node does not receive an RA, and decides to transmit an RS, a broadcast Aloha slot (see ) is consumed with unencrypted traffic. already allows for a unicast reply to such an RS. This is a particularly acute issue for the join process for the following reasons:

1. Use of a multicast slot by even a non-malicious unauthenticated node for a Router Solicitation (RS) may overwhelm that timeslot.
2. It may require many seconds of on-time before a new pledge receives a Router Advertisement (RA) that it can use.
3. A new pledge may have to receive many EBs before it can pick an appropriate network and/or closest Join Proxy to attach to. If it must remain in the receive state for an RA as well as find the EB, then the process may take dozens of seconds, even minutes for each enrollment attempt that it needs to make.

Layer 2 Selection In a complex Low-power and Lossy Network (LLN), multiple LLNs may be connected together by Backbone Routers (technology such as ), resulting in an area that is serviced by multiple, distinct Layer 2 instances. These are called Personal Area Networks (PANs). Each instance will have a separate Layer 2 security profile and will be distinguished by a different PANID. The PANID is part of the Layer 2 header as defined in : it is a 16-bit value that is chosen to be unique, and it contributes context to the Layer 2 security mechanisms. The PANID provides a context similar to the Extended Service Set ID (ESSID) in 802.11 networking and can be considered similar to the 802.3 Ethernet VLAN tag in that it provides context for all Layer 2 addresses. A device that is already enrolled in a network may find after a long sleep that it needs to resynchronize with the Layer 2 network. The device's enrollment keys will be specific to a PANID, but the device may have more than one set of keys. Such a device may wish to connect to a PAN that is experiencing less congestion or that has a shallower Routing Protocol for LLNs (RPL) tree . It may even observe PANs for which it does not have keys, but for which it believes it may have credentials that would allow it to join. In order to identify which PANs are part of the same backbone network, the network ID is introduced in this extension. PANs that are part of the same backbone will be configured to use the same network ID. For RPL networks , configuration of the network ID can be done with a configuration option, which is the subject of future work. In order to provide some input to the choice of which PAN to use, the PAN priority field has been added. This lists the relative priority for the PAN among different PANs. Every EB from a given PAN will likely have the same PAN priority. Determination of the PAN priority is the subject of future work; but it is expected that it will be calculated by an algorithm in the 6LoWPAN Border Router (6LBR), possibly involving communication between 6LBRs over the backbone network. The parent selection process can only operate within a single PAN because it depends upon receiving RPL DIO messages from all available parents. As part of the PAN selection process, the device may wish to know how deep in the LLN mesh it will be if it joins a particular PAN, and the rank priority field provides an estimation of each announcer's rank. Once the device synchronizes with a particular PAN's TSCH schedule, it may receive DIOs that are richer in their diversity than this value. The use of this value in practice is the subject of future research, and the interpretation of this value of the structure is considered experimental.

Protocol Definition creates a registry for new IETF IE subtypes. This document allocates a new subtype. The new IE subtype structure is as follows. As explained in , the length of the subtype content can be calculated from the container, so no length information is necessary.

IE Subtype Structure

res:

Reserved bits MUST be ignored upon receipt and SHOULD be set to 0 when sending.

R:

The RA R-flag is set if the sending node will act as a router for host-only nodes relying on stateless address auto-configuration (SLAAC) to get their global IPv6 address. Those hosts MUST send a unicast RS message in order to receive an RA with the Prefix Information Option.

In most cases, every node sending a beacon will set this flag, and in a typical mesh, this will be every single node. When this bit is not set, it might indicate that this node may be under provisioned or that it may have no additional slots for additional nodes. This could make this node more interesting to an attacker.

P:

If the Proxy Address P-flag is set, then the Join Proxy Interface ID bit field is present. Otherwise, it is not provided.

This bit only indicates if another part of the structure is present, and it has little security or privacy impact.

proxy prio (proxy priority):

This field indicates the willingness of the sender to act as a Join Proxy. Lower value indicates greater willingness to act as a Join Proxy as described in . Values range from 0x00 (most willing) to 0x7e (least willing). A priority of 0x7f indicates that the announcer should never be considered as a viable Join Proxy. Only unenrolled pledges look at this value.

Lower values in this field indicate that the transmitter may have more capacity to handle unencrypted traffic. A higher value may indicate that the transmitter is low on neighbor cache entries or other resources. Ongoing work such as documents one way to set this field.

rank priority:

The rank priority is set by the IPv6 LLN Router (6LR) that sent the beacon and is an indication of how willing this 6LR is to serve as a RPL parent within a particular network ID. Lower values indicate more willingness, and higher values indicate less willingness. This value is calculated by each 6LR according to algorithms specific to the routing metrics used by the RPL . The exact process is a subject of significant research work. It will typically be calculated from the RPL rank, and it may include some modifications based upon current number of children or the number of neighbor cache entries available. Pledges MUST ignore this value. It helps enrolled devices only to compare connection points.

An attacker can use this value to determine which nodes are potentially more interesting. Nodes that are less willing to be parents likely have more traffic, and an attacker could use this information to determine which nodes would be more interesting to attack or disrupt.

PAN priority:

The PAN priority is a value set by the Destination-Oriented Directed Acyclic Graph (DODAG) root (see , typically the 6LBR) to indicate the relative priority of this LLN compared to those with different PANIDs that the operator might control. This value may be used as part of the enrollment priority, but typically it is used by devices that have already enrolled and need to determine which PAN to pick when resuming from a long sleep. Unenrolled pledges MAY consider this value when selecting a PAN to join. Enrolled devices MAY consider this value when looking for an eligible parent device. Lower values indicate more willingness to accept new nodes.

An attacker can use this value, along with the observed PANID in the EB to determine which PANIDs have more network resources, and may have more interesting traffic.

Join Proxy Interface ID:

If the P bit is set, then 64 bits (8 bytes) of address are present. This field provides the Interface ID (IID) of the link-local address of the Join Proxy. The associated prefix is well-known as fe80::/64. If this field is not present, then IID is derived from the Layer 2 address of the sender per SLAAC .

This field communicates the IID bits that should be used for this node's Layer 3 address, if it should not be derived from the Layer 2 address. Communication with the Join Proxy occurs in the clear. This field avoids the need for an additional service-discovery process for the case where the Layer 3 address is not derived from the Layer 2 address. An attacker will see both Layer 2 and Layer 3 addresses, so this field provides no new information.

network ID:

This is a variable length field, up to 16-bytes in size that uniquely identifies this network, potentially among many networks that are operating in the same frequencies in overlapping physical space. The length of this field can be calculated as being whatever is left in the IE.

In a 6TiSCH network, where RPL is used as the mesh routing protocol, the network ID can be constructed from a truncated SHA-256 hash of the prefix (/64) of the network. This will be done by the RPL DODAG root and communicated by the RPL Configuration Option payloads, so it is not calculated more than once. This is just a suggestion for a default algorithm: it may be set in any convenient way that results in a non-identifying value. In some LLNs where multiple PANIDs may lead to the same management device (the Join Registrar/Coordinator (JRC)), then a common value that is the same across all the PANs MUST be configured. Pledges that see the same network ID will not waste time attempting to enroll multiple times with the same network when the network has multiple attachment points.

If the network ID is derived as suggested, then it will be an opaque, seemingly random value and will not directly reveal any information about the network. An attacker can match this value across many transmissions to map the extent of a network beyond what the PANID might already provide.

Security Considerations All of the contents of this IE are transmitted in the clear. The content of the EB is not encrypted. This is a restriction in the cryptographic architecture of the 802.15.4 mechanism. In order to decrypt or do integrity checking of Layer 2 frames in TSCH, the TSCH ASN is needed. The EB provides the ASN to new (and long-sleeping) nodes. The sensitivity of each field is described within the description of each field. The EB is authenticated at the Layer 2 level using 802.15.4 mechanisms using the network-wide keying material. Nodes that are enrolled will have the network-wide keying material and can validate the beacon. Pledges that have not yet enrolled are unable to authenticate the beacons and will be forced to temporarily take the contents on faith. After enrollment, a newly enrolled node will be able to return to the beacon and validate it. In addition to the enrollment and join information described in this document, the EB contains a description of the TSCH schedule to be used by the transmitter of this packet. The schedule can provide an attacker with a list of channels and frequencies on which communication will occur. Knowledge of this can help an attacker to more efficiently jam communications, although there is future work being considered to make some of the schedule less visible. Encrypting the schedule does not prevent an attacker from jamming, but rather increases the energy cost of doing that jamming.

Privacy Considerations The use of a network ID may reveal information about the network. The use of a SHA-256 hash of the DODAGID (see ), rather than using the DODAGID itself directly provides some privacy for the addresses used within the network, as the DODAGID is usually the IPv6 address of the root of the RPL mesh. An interloper with a radio sniffer would be able to use the network ID to map out the extent of the mesh network.

IANA Considerations IANA has assigned the following value in the "IEEE Std 802.15.4 IETF IE Subtype IDs" registry, as defined by .

Value	Subtype ID	Reference
2	6tisch-Join-Info	RFC 9032

References Normative References IEEE Informative References

Acknowledgments provided extensive editorial comments on the document. generated a detailed review of the document at Working Group Last Call. provided a number of useful editorial suggestions.